English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Deep generative models for structure-based drug design (SBDD), where moleculegeneration is conditioned on a 3D protein pocket, have received considerableinterest in recent years. These methods offer the promise of higher-qualitymolecule generation by explicitly modelling the 3D interaction between apotential drug and a protein receptor. However, previous work has primarilyfocused on the quality of the generated molecules themselves, with limitedevaluation of the 3D molecule \emph{poses} that these methods produce, withmost work simply discarding the generated pose and only reporting a "corrected"pose after redocking with traditional methods. Little is known about whethergenerated molecules satisfy known physical constraints for binding and theextent to which redocking alters the generated interactions. We introducePoseCheck, an extensive analysis of multiple state-of-the-art methods and findthat generated molecules have significantly more physical violations and fewerkey interactions compared to baselines, calling into question the implicitassumption that providing rich 3D structure information improves moleculecomplementarity. We make recommendations for future research tacklingidentified failure modes and hope our benchmark can serve as a springboard forfuture SBDD generative modelling work to have a real-world impact.

Benchmarking Generated Poses: How Rational is Structure-based Drug Design with Generative Models?